Background
The authors write that "effects of global warming on streamflow magnitude are of particular concern given that these effects directly alter water supplies." In this regard, however, they say that "previous studies of trends and variability of streamflow in the United States primarily have focused on the latter half of the 20th century," noting that these trends "may not be representative of longer time periods." Thus, they decided to significantly increase the time domain of their new study of the subject to cover the period 1900-2008, stating that it would "be informative to evaluate temporal patterns in streamflow over the longest possible period," which is critically important, in light of the fact that earth's recovery from the global chill of the Little Ice Age began well over a century ago.

What was done
Working with monthly temperature and precipitation data provided on a 4 km x 4 km grid that they obtained from the Parameter-elevation Regression on Independent Slopes Model or PRISM data set, coupled with their use of the water-balance model previously employed by McCabe and Wolock (1999) and Wolock and McCabe (1999), the two U.S. Geological Survey researchers examined the separate effects of temperature and precipitation on U.S. runoff variability from 1900 to 2008 to see if they could find a signature of global warming.

What was learned
McCabe and Wolock determined, as they describe it, that over the past century "precipitation accounts for almost all of the variability (98%) in mean water-year United States runoff," noting that, in contrast, temperature effects on runoff have been small for most locations in the country, "even during periods when temperatures for most of the United States increased significantly." They do find a small increase in runoff over the course of their record, however, due to a distinct increase that occurred around 1970. But they suggest that the small but abrupt change may be due to a shift in the North Atlantic Oscillation (NAO), since, in their words, "precipitation in the eastern United States is positively correlated with the NAO (Hurrell, 1995), and around 1970 the NAO shifted from a primarily negative phase to a primarily positive phase (Milly and Dunne, 2001)," and they note that "when the NAO is in a positive phase, southerly winds over the eastern United States are strengthened and enhance the transport of moisture from the Gulf of Mexico into the eastern United States," which then "results in increased precipitation and runoff (Milly and Dunne, 2001)."

What it means
In light of their comprehensive analysis and subsequent findings, McCabe and Wolock state that "because precipitation variability controls runoff variability, little confidence can be placed in climate-model estimates of future runoff without reliable estimates of precipitation."

References
Hurrell, J.W. 1995. Decadal trends in the North Atlantic Oscillation: Regional temperatures and precipitation. Science 269: 676-679.

McCabe, G.J. and Wolock, D.M. 1999. Future snowpack conditions in the western United States derived from general circulation model climate simulations. Journal of the American Water Resources Association 35: 1473-1484.

Milly, P.C.D. and Dunne, K.A. 2001. Trends in evaporation and surface cooling in the Mississippi river basin. Geophysical Research Letters 28: 1219-1222.

Wolock, D.M. and McCabe, G.J. 1999. Effects of potential climatic change on annual runoff in the conterminous United States. Journal of the American Water Resources Association 35: 1341-1350.